[Ivan TT]

By the way, what zobel config are you using.

Tested Zobel network, 55Ohm 5uF (should be 5.5 but tricky to source) on dual RAB in series, 100Ohm (should be 110Ohm) and 2.5uF on quad in series, arranged on breadborad so that I can bring Zobel in and out using two wires.
Results are... inconclusive and strange.
The effect on dual in series is subtle, but noticeable: wider more 3D soundstage, more natural convincing sound overall and quite improved bass extension (taking it into almost CI territory), on quad it's more subtle (including 55Ohm/5uF configuration). No effect on Bellsing 5 at all (tested out of curiosity).
I'm not sure if I like it or not, but the difference is there, even if ever so subtle.
Will need to listen to the effect more.

 

[dhruvmeena96]

Tested Zobel network, 55Ohm 5uF (should be 5.5 but tricky to source) on dual RAB in series, 100Ohm (should be 110Ohm) and 2.5uF on quad in series, arranged on breadborad so that I can bring Zobel in and out using two wires.
Results are... inconclusive and strange.
The effect on dual in series is subtle, but noticeable: wider more 3D soundstage, more natural convincing sound overall and quite improved bass extension (taking it into almost CI territory), on quad it's more subtle (including 55Ohm/5uF configuration). No effect on Bellsing 5 at all (tested out of curiosity).
I'm not sure if I like it or not, but the difference is there, even if ever so subtle.
Will need to listen to the effect more.

Quad RAB32257 500Hz standard zobel circuit is
Cz= 1.2 uF
Rz= 165ohms

Dual series RAB32257 500Hz standard zobel is
Cz=2.3uF
Rz=82.5ohms



Since you calculated from DCR standard

(those value are from DCR standard)

i think Bass effect would be more as the RAB will roll of earlier compared to 500Hz standard
Yours would have more soundstage as the driver is now looking more like DCR in most of response, thus reducing electrical and signal distortion, plus now the roll of treble and mids turn things more vocal and fundamental tone oriented, and zobel eats up sensitivity, distortion and ground floor, things starts to sound wider and deeper

inconclusive and strange right

if you find the bass too much, then you can have the Zobel made from 500Hz standard

 

[dhruvmeena96]

Well,
RAB32063 driver with Zobel(DCR standard) and 40ohms series resistor(after zobel, going toward connector for iem) with 10mm lenght/2mm ID tubing, white damper(680 ohms) at 3mm from driver spout

Cz = 5 uF(ideal 4.95 uF)
Rz = 38 ohms(ideal 38.125 ohms)

Rs = 40ohms

pure listening bliss.
Smoothest treble, fantastic mids and way more controlled bass amount and dynamics.
Linear Bass response with better extension

DCR = 62 ohms
Impedance = 86.5 ohms

@Ivan TT
Higher impedance load perform better on voltage source amp in internal driver distortion which gives them black sound, but distorts amps slightly to give the texture and grain. i was listening to Quad RAB32257 series, and it sounded like when you add good grain while editing videos. that grain hide source artifacts and sharpen image, the same thing happened to Quad series. low quality 128kbps Eminem Lose yourself sounded tolerable, and when i shifted to FLAC on the same music, it subdued the old microphone hiss and made instruments stand out. That was like revelation to me, because i thought series were bad as a configuration. But the bass seems strong AF when powered by O2(not joking, it is stronger than TFZ Queen and Exclusive KING).

 

[dhruvmeena96]

This is my try to diy iem,
Using rab 30233+wbfk 30095 +ed 29689 all in series..
No cross over, all driver work in fullrange

The rab and wbfk being modified by take off the front sound tunnel

I heard this sound balanced from. Mid bass to high

Any idea to. Enhance the sound?

Thanks in advance

Put a crossover on WBFK...a capacitor

And make it parallel to ED and RAB

 

[Ivan TT]

So after spending literally HOURS evaluating Zobels (actually Boucherot cells), I can report preliminary findings/impressions:

The effect is definitely there and has 3 main aspects to it:
1. High frequency quality change: I observe subtle uplift, extra extension (noticeable on cymbals) and somewhat silkier quality. Also vocals change presence-positioning quite a bit. Overall sounds like very nice "presence" control turned on. Resistor value seems to have most effect on this, best results are around x1.25 DC resistance or impedance @ 500Hz
3. Soundstage: wider and deeper, sometimes a touch too deep (I think maybe due to phase shift introduced by Zobel?)
4. Low end - extra extension, not louder but deeper and more bass authority. Seems to be slightly more pronounced when higher capacitance is used, 2.2-4.7uF is a sweet spot for me.

Best effect so far was observed on x4 RAB in series, which was a bit underwhelming but now sounds great!
x4 RAF - definitely an improvement, but with Zobel it is not as good as RAB-Z (Z for Zobel).
x2 RAB-Z - very nice, maybe a touch too much highs, I'm tempted to try red damper to see what happens (currently green).
Bellsing 5 - I can JUST hear the effect (or can convince myself I do).

Overall introduction of Zobel improves BAs sound within that elusive 5% (I had more radical or comparable change in sound signature from tips/cable rolling or changing dampers), but also proves tricky to fine-tune due to how subtle it is (and changes between different component values are even subtler), I would recommend using standard formula but doubling the capacitor value.

 

[dhruvmeena96]

So after spending literally HOURS evaluating Zobels (actually Boucherot cells), I can report preliminary findings/impressions:

The effect is definitely there and has 3 main aspects to it:
1. High frequency quality change: I observe subtle uplift, extra extension (noticeable on cymbals) and somewhat silkier quality. Also vocals change presence-positioning quite a bit. Overall sounds like very nice "presence" control turned on. Resistor value seems to have most effect on this, best results are around x1.25 DC resistance or impedance @ 500Hz
3. Soundstage: wider and deeper, sometimes a touch too deep (I think maybe due to phase shift introduced by Zobel?)
4. Low end - extra extension, not louder but deeper and more bass authority. Seems to be slightly more pronounced when higher capacitance is used, 2.2-4.7uF is a sweet spot for me.

Best effect so far was observed on x4 RAB in series, which was a bit underwhelming but now sounds great!
x4 RAF - definitely an improvement, but with Zobel it is not as good as RAB-Z (Z for Zobel).
x2 RAB-Z - very nice, maybe a touch too much highs, I'm tempted to try red damper to see what happens (currently green).
Bellsing 5 - I can JUST hear the effect (or can convince myself I do).

Overall introduction of Zobel improves BAs sound within that elusive 5% (I had more radical or comparable change in sound signature from tips/cable rolling or changing dampers), but also proves tricky to fine-tune due to how subtle it is (and changes between different component values are even subtler), I would recommend using standard formula but doubling the capacitor value.

doubling the Cz of zobel rolls of the driver earlier, damping 2kHz peak and other peak. Standard zobel only dampens the rising impedance and in my experience takes away the distortion and roughness in treble.

When i first talked about zobel, i said something about increase in depth information.
Well, that happens because when HF is electrically rolled of while not having anything in series to circuit, makes the driver distortion flatter and lower on treble plus decreasing the load of air resistance to diaphragm, which tends to clean the 1kHz spike in BA and lowering bass, bass and treble induced distortions to the mids. Second, zobel waste the energy of a circuit while keeping most of the dB same. so take pink or white noise and just remove the HF part, the noise become easier to listen and slightly pleasing, zobel does the same too. Floor noise is lower. This overall contributes to the depth information and sound-stage even though treble is technically lower(so called air)

Bass increase is due to de-loading HF in roll off pattern from driver, giving it breather and helping it go deeper in Bass.

treble issues happens as zobel don't kill the peaks, they remain the same as it is. so the linear roll off makes it more noticeable in a response. a damper rolls off plus dampens the peak.

hmmmm
decreasing resistor decrease treble
increasing resistor increases treble to a limit and then decrease it back again




nowadays
i calculate the zobel for whole impedance range for the given inductance and then average it out.
thats the perfect value for me

 

[Kulgrinda]

So after spending literally HOURS evaluating Zobels (actually Boucherot cells), I can report preliminary findings/impressions:

The effect is definitely there and has 3 main aspects to it:
1. High frequency quality change: I observe subtle uplift, extra extension (noticeable on cymbals) and somewhat silkier quality. Also vocals change presence-positioning quite a bit. Overall sounds like very nice "presence" control turned on. Resistor value seems to have most effect on this, best results are around x1.25 DC resistance or impedance @ 500Hz
3. Soundstage: wider and deeper, sometimes a touch too deep (I think maybe due to phase shift introduced by Zobel?)
4. Low end - extra extension, not louder but deeper and more bass authority. Seems to be slightly more pronounced when higher capacitance is used, 2.2-4.7uF is a sweet spot for me.

Best effect so far was observed on x4 RAB in series, which was a bit underwhelming but now sounds great!
x4 RAF - definitely an improvement, but with Zobel it is not as good as RAB-Z (Z for Zobel).
x2 RAB-Z - very nice, maybe a touch too much highs, I'm tempted to try red damper to see what happens (currently green).
Bellsing 5 - I can JUST hear the effect (or can convince myself I do).

Overall introduction of Zobel improves BAs sound within that elusive 5% (I had more radical or comparable change in sound signature from tips/cable rolling or changing dampers), but also proves tricky to fine-tune due to how subtle it is (and changes between different component values are even subtler), I would recommend using standard formula but doubling the capacitor value.

Thanks for sharing! Can you give R and C values for 2xRAB and 4xRAB setups?

 

[Ivan TT]

Thanks for sharing! Can you give R and C values for 2xRAB and 4xRAB setups?

The pleasure is mine
Actual values are the same for RAB and RAF x2 and are a bit tricky as @dhruvmeena96 points out above there's DC resistance and impedances @ different Hz, if one to take impedance of 33Ohm @500Hz as a base value it's 82.5Ohm and 2.3uF (truncated to the closest values of components you can source, most likely 82Ohm and 2.2uF)
I would also try x2 capacitor value (4.7uF or 2x2.2uF in parallel), it gives slightly fuller sound unless I imagine it.

Spoiler: This is where things get interesting

If you want to dig deeper and DIY:
Calculator:
http://diyaudioprojects.com/Technical/Speaker-Zobel/
Datasheets:
https://www.knowles.com/docs/default-source/model-downloads/rab-32257-000.pdf?Status=Master&sfvrsn=0
https://www.knowles.com/docs/defaul...dbb37cff0000940c19.pdf?Status=Master&sfvrsn=0
Any value (DC resistance and impedances @ different Hz) could be a starting point and are worth testing if you have time/components) but I notice that their average value as @dhruvmeena96 suggests is close to impedance @500 Hz

 

[dhruvmeena96]

The pleasure is mine
Actual values are the same for RAB and RAF x2 and are a bit tricky as @dhruvmeena96 points out above there's DC resistance and impedances @ different Hz, if one to take impedance of 33Ohm @500Hz as a base value it's 82.5Ohm and 2.3uF (truncated to the closest values of components you can source, most likely 82Ohm and 2.2uF)
I would also try x2 capacitor value (4.7uF or 2x2.2uF in parallel), it gives slightly fuller sound unless I imagine it.

Spoiler: This is where things get interesting

If you want to dig deeper and DIY:
Calculator:
http://diyaudioprojects.com/Technical/Speaker-Zobel/
Datasheets:
https://www.knowles.com/docs/default-source/model-downloads/rab-32257-000.pdf?Status=Master&sfvrsn=0
https://www.knowles.com/docs/defaul...dbb37cff0000940c19.pdf?Status=Master&sfvrsn=0
Any value (DC resistance and impedances @ different Hz) could be a starting point and are worth testing if you have time/components) but I notice that their average value as @dhruvmeena96 suggests is close to impedance @500 Hz

Its close to 500Hz but has lower damping factor(higher resistance, but at same time, is overall more energy dissipator compared to 500HZ and also sounds better too)

Good first build for RAB user

RAB32063(non vented) + Zobel(500Hz) + L-pad(with more series added)

L-pad config
Rp = 10ohms/ 1Watt
Rs = 40ohms/ 1watt

8mm lenght / 1mm ID tube
Brown damper at center

Fastest and the least distorting BA setup.
500mW rated safe(1W should never be thrown, as it will blow the resistor in prolonged use) which is high. The bass linearity was the most shocking with the mids, plus it doesnt get lost like etymotic BA roll off. Very extended and speedy transition from single BA

 

[zekester]

Thanks for this epic thread! I have read huge chunks at a time, made and failed, made and succeeded in making shells, wired up and created my first GV based set and used them on stage. This has been an incredibly fun undertaking and thank you for all of your years of activity to make it happen.

I have read through and researched pretty much every aspect of the process and have many questions. The first one though is how do you determine the effect of tubing length? I understand that length and diameter has an effect on the tuning on several levels, but it seems like phase and a frequency filtering effect both take place. Is there a calculation that shows rolloff at certain lengths for example?

Thanks again, I am hooked!

 

[eunice]

Some in this thread measure the effect by starting with a long tube and cutting Millimeter by millimeter. Others ( @dhruvmeena96 ) use expensive physics simulation software to solve it.

I for one did not (yet) bother, I try to make the tubes as short as possible, I have not yet heard any improvement by making some tubes longer in my sets. However I do hear if there is a short piece of 1mm ID tube inside the 2mm ID tube at the driver.

So good question, would be interested in info as well, did someone come up with a few rules of thumb?

 

[dhruvmeena96]

Some in this thread measure the effect by starting with a long tube and cutting Millimeter by millimeter. Others ( @dhruvmeena96 ) use expensive physics simulation software to solve it.

I for one did not (yet) bother, I try to make the tubes as short as possible, I have not yet heard any improvement by making some tubes longer in my sets. However I do hear if there is a short piece of 1mm ID tube inside the 2mm ID tube at the driver.

So good question, would be interested in info as well, did someone come up with a few rules of thumb?

Smaller ID shifts peak backward while also damping it.
Bigger ID retain the peak position but also dampen(slightly).
Reference ID for Knowles = 2mm
Reference ID for Sonion = 2mm/1.5mm

Longer length tubing act as a low pass

For example
If 2mm ID / 10mm lenght is reference on RAB driver

Then changing 2mm to 1.5mm or 1mm will shift its inherent peak 3kHz toward 2kHz while also shifting other peaks and damping it.

If length is increased to 20mm, then it will go through a longer tube, loosing the HF as longer tube will dissipate HF energy plus more distance of driver to eardrum.

Its all in combination of ID and Length.

Read sonion and Knowles guide for more info

 

[dhruvmeena96]

Well, does anybody have measurements of Sonion 2354 vs 2389 and ED29689
well i am interested in 2354 vs ED29689

plus Sonion 2354 is 25$
i dont know if it is pair or single driver

https://www.yoycart.com/Product/559474794045/

 

[Xymordos]

Not sure if you got this before, there's measurements of the 29689 driver (and some other Knowles ones) here.

Edit: seems like you can't upload files, I've uploaded it here: https://pan.baidu.com/s/18qTpYqhLiKcdbw47_ECxxg

PW: 8m9x

 

[NoForMe]

I've searched and searched, I feel like the last thing I need is a colloid dispenser. Has anyone found anything that makes this easier than heating in a microwave or stove? The dispensers I have found run well over $800. If someone has a DIY hack please let me know!